LAUSR

In this paper, we propose an effective image fusion technique employing a deep learning framework. The novelty of the project is to process image fusion to collect utmost information that is potential. We tend to commit to use the architecture to extract a lot of useful features from source images within the encryption method. Image fusion of infrared (IR) image and visible image (VI) can combine the advantages of thermal radiation information in infrared images and detailed texture information in visible images. Military, navigation, medical analysis, remote sensing, photography, and weapon detection would like completely different imaging modalities such as visible and infrared to monitor a targeted scene. These imaging modalities provide complementary information. This complementary information of 'n' variety of images is integrated to create one image. This process is known as Image Fusion. Since most of the images while capturing are affected by sure conditions like poor lighting, or weather, the images suffer from poor distinction. Thus, we further apply a computer image processing technique for enhancement of fused image. Therefore, improving the standard of images for human observer is done. In this paper, we used convolution neural network (CNN) for fusing IR and VI images. CNN is a deep learning-based technique, which can extract features efficiently than existing techniques. Our methodology is tested on many image pairs and is evaluated qualitatively by visual inspection and quantitatively. The performing fusion method accurately integrates the target information from IR and background details of VI. Qualitative and quantitative results support that the proposed technique is better than the existing techniques.